We commonly think of chemical mutation as a set of DNA adducts which are either premutagenic lesions, inducers of cellular response systems, or both. We propose to develop a general technology to test this proposition. We intend to map the position, count the number and, ultimately, identify the chemical structure of sequence specific DNA adducts formed in experiments in which mutation is also studied as a function of DNA sequence. Our strategy is to first separate adducted DNA sequences by denaturing gradient gel electrophoresis. Secondly, we propose to identify the position of adducted DNA sequences on the gel by DNA amplification of individual gel slices. Finally, we propose to map the positions of the adducts in the identified DNA sequences by use of DNA polymerase or exonuclease blocking or, in some cases, by mapping mutations arising during DNA amplification. We propose to focus on the mutagens MNNG and H2O2 to support the efforts of our Program collaborators in the study of alkylating and oxidizing agent induced mutation and because the expected adducts from these chemicals will challenge our analytical capability. In particular, we propose to perform an input-output analysis, adduct spectra map and mutation spectra map, for MNNG and H2O2 in diploid human cells. If we can reach these goals, we further propose to develop appropriate technology to provide sufficient site specific DNA adduct material for direct study by mass spectrometry which we estimate to be 1012to1013 molecules.